{"id":1977,"date":"2026-08-26T10:23:50","date_gmt":"2026-08-26T10:23:50","guid":{"rendered":"https:\/\/blog.openzeka.com\/en\/?p=1977"},"modified":"2026-08-26T10:23:50","modified_gmt":"2026-08-26T10:23:50","slug":"qwen3-8-27b-on-dgx-spark-48-tok-s","status":"publish","type":"post","link":"https:\/\/blog.openzeka.com\/en\/qwen3-8-27b-on-dgx-spark-48-tok-s\/","title":{"rendered":"Qwen3.8-27B on DGX Spark \u2014 48 tok\/s"},"content":{"rendered":"

Qwen3.8-27B is a 27-billion-parameter language model released by Alibaba in August 2026. With a score of 52 on the Artificial Analysis Intelligence Index, it ranks first in its class among open-weight models under 40B parameters.<\/p>\n

In this tutorial, you will run Qwen3.8-27B-NVFP4 on a single DGX Spark using the SGLang inference engine and the DFlash2 draft model for speculative decoding, reaching a generation speed of 48 tokens per second. SGLang loads the model’s trained weights into GPU memory and exposes an API that accepts external requests. sparkrun manages the Docker container and automates distribution of the model files.<\/p>\n

The Spark can be used directly as a computer by connecting a monitor and keyboard, or it can operate as a server accessed remotely from another computer. In this tutorial, we will connect to the Spark remotely, set up the required software, and launch the model.<\/p>\n

This tutorial is divided into four sections:<\/p>\n

    \n
  • Setup:<\/strong> Pulling the Docker image and preparing the recipe<\/li>\n
  • Running:<\/strong> Starting, monitoring, and testing the model with sparkrun<\/li>\n
  • Benchmark:<\/strong> Performance measurements at different concurrency levels<\/li>\n
  • Shutdown:<\/strong> Stopping the services<\/li>\n<\/ul>\n

    Setup<\/h2>\n

    1. Connecting to the Spark<\/h3>\n

    If this is your first time connecting to the Spark remotely, you first need to find its IP address. Connect a monitor and keyboard to the Spark, log in, and run the following command in a terminal:<\/p>\n<\/div>